Method and system for optimizing low-speed starting noise of permanent magnet synchronous motor blower

Abstract

The invention discloses a method and system for optimizing low-speed starting noise of a permanent magnet synchronous motor blower, and the method comprises the steps: judging the starting rotating speed of a motor, executing transient initial positioning of a rotor at zero speed, abandoning a conventional open-loop forced dragging stage, and directly starting a full-order state observer based on a motor mathematical model to enter closed-loop control. The observer works under a synchronous rotating coordinate system, the angular speed and angle of a rotor are estimated in real time according to a voltage equation, and rapid convergence near the zero speed is ensured by setting the lower limit of the angular speed. Electromagnetic noise and high-frequency howling generated by pre-positioning and open-loop strong dragging are eliminated fundamentally, the technical problem that a counter electromotive force observation method fails due to weak signals at the extremely low speed is solved, and smooth and mute starting of the air blower from the zero speed is achieved. Meanwhile, the system is compatible with a single / double-resistor sampling scheme, and has the advantages of high performance and low cost.

Description

Technical Field

[0001] This invention relates to the field of motor control technology, specifically to a sensorless start-up control method and system for a permanent magnet synchronous motor (PMSM), which is particularly suitable for devices such as vehicle air conditioning blowers and cooling fans that have strict requirements for start-up noise. Background Technology

[0002] As the automotive industry undergoes a profound transformation towards electrification and intelligentization, the widespread adoption of new energy vehicles has eliminated engine noise, the primary noise source in traditional gasoline-powered vehicles. Simultaneously, tire noise reduction and wind noise control technologies are becoming increasingly sophisticated. Against this backdrop, previously masked auxiliary system noise, particularly the operating noise of the blower motor in HVAC (heating, ventilation, and air conditioning) systems, has become a key factor affecting driving quietness and comfort. The noise problem of the motor during low-speed startup is especially prominent.

[0003] Currently, for surface-mounted or built-in permanent magnet synchronous motors (PMSMs), the sensorless low-speed start-up generally adopts the classic "three-stage" control strategy: from pre-alignment to open-loop I / F control and then to closed-loop FOC.

[0004] During the pre-positioning phase, the controller applies a fixed DC voltage vector to the motor stator windings, forcibly pulling the rotor to a known initial angle. This process generates significant instantaneous torque and current surges, resulting in noticeable "humming" electromagnetic noise.

[0005] During the open-loop forced acceleration phase, the controller ignores the actual position of the rotor and increases the output frequency according to the preset voltage-frequency (V / F) curve to forcibly accelerate the motor in an open-loop manner. Due to the continuous error between the applied voltage vector and the actual position of the rotor, significant torque pulsation and current harmonics will be caused, which in turn will excite mechanical resonance between the motor body and the load, producing an uncomfortable high-frequency "whistling" sound and vibration.

[0006] The system switches to closed-loop field-oriented control (FOC) based on back EMF observation only after the motor speed increases sufficiently to generate a detectable back electromotive force (BEMF). However, in the extremely low-speed region (typically below 5% of rated speed), the back EMF amplitude is extremely small, resulting in a very low signal-to-noise ratio (SNR), leading to large observer estimation errors, slow convergence speed, and even instability. This often causes the switching process from open-loop to closed-loop control to be accompanied by speed and torque surges, generating additional noise.

[0007] Furthermore, at the hardware level, there is a trade-off between single-resistor and dual-resistor current sampling schemes. While the dual-resistor scheme simplifies current reconstruction, it is more expensive; the single-resistor scheme is inexpensive, but requires multiple precise samples in each PWM cycle and relies on complex reconstruction algorithms, placing high demands on the computational power of the microcontroller (MCU) and easily introducing reconstruction errors.

[0008] In summary, the existing technology has the following inherent defects:

[0009] 1) Inherent noise sources in the startup process (pre-positioning impact, open-loop torque pulsation);

[0010] 2) Insufficient low-speed observation capability (back EMF method fails at low speeds);

[0011] 3) It is difficult to balance performance and cost.

[0012] Therefore, there is an urgent need in this field for an innovative control scheme that can optimize the startup process in principle and achieve a smooth and quiet startup at extremely low speeds. Summary of the Invention

[0013] The primary objective of this invention is to overcome the shortcomings of the prior art and provide a method and system for optimizing the low-speed start-up noise of a permanent magnet synchronous motor blower. This method aims to fundamentally eliminate the two main noise-generating stages of pre-positioning and open-loop forced drag, and solve the problem of accurate rotor position observation at extremely low speeds, thereby achieving smooth, quiet, and reliable closed-loop start-up from zero speed.

[0014] To achieve the above objectives, the present invention provides the following technical solution:

[0015] A method for optimizing the low-speed start-up noise of a permanent magnet synchronous motor blower, characterized by comprising the following steps:

[0016] In response to the start command, determine whether the current speed of the permanent magnet synchronous motor is zero;

[0017] If the current rotational speed is zero, perform an initial rotor positioning operation.

[0018] After the initial rotor positioning operation, without going through the open-loop strong drag stage, the full-order state observer based on the motor model is directly activated for closed-loop control.

[0019] The full-order state observer is based on the motor voltage equation in the synchronous rotating coordinate system. It estimates the rotor's angular velocity and angle in real time according to the input voltage signal and the sampled current signal, and outputs the results to the field-oriented control (FOC) loop.

[0020] Compared to the traditional three-stage start-up method of "pre-positioning - open-loop forced drag - closed-loop", this invention eliminates the open-loop forced drag stage and directly switches to closed-loop control after initial positioning. This fundamentally eliminates the torque pulsation and high-frequency electromagnetic noise caused by position estimation errors in open-loop forced drag, and achieves root-cause noise reduction and stability improvement in the start-up process.

[0021] Preferably, the rotor initial positioning operation is as follows: injecting a DC current vector of preset magnitude and direction into the stator winding of the motor, and using the magnetoresistive effect to align the rotor permanent magnet to a determined initial position.

[0022] By limiting the initial positioning to injecting a preset DC current vector, compared with the traditional positioning method that may use a larger current or voltage, a shorter and gentler positioning is achieved, which significantly reduces the current surge and accompanying electromagnetic noise in the initial positioning stage, and provides optimized start-up conditions for subsequent observers.

[0023] Preferably, the full-order state observer can achieve real-time angle convergence within 2 electrical cycles after the rotor initial positioning operation is completed, thereby realizing full-speed domain angle closed-loop control.

[0024] By setting a lower limit for the angular velocity estimated by the full-order state observer, the theoretical problem of poor observability and difficulty in starting the observer near zero velocity is effectively solved, ensuring the rapid convergence and stability of the observer at the moment of startup, and providing a key guarantee for achieving silent and rapid startup.

[0025] Preferably, the full-order state observer is able to converge and accurately identify the real-time angle of the rotor within two electrical cycles after the rotor initial positioning operation is completed.

[0026] By limiting the observer to converge rapidly within 2 electrical cycles, the present invention has achieved an order-of-magnitude improvement in response speed compared to the traditional open-loop strong drag process that requires tens or even hundreds of electrical cycles. This rapid locking capability directly brings about a qualitative leap in start-up smoothness and noise reduction.

[0027] More preferably, the method further includes:

[0028] If the current speed is not zero, or when the motor operating speed is higher than a preset threshold, switch to closed-loop control based on the back EMF observer.

[0029] By adding a smart switching strategy for high and low speed observers, the control system not only solves the problem of low-speed start-up noise, but also leverages the advantages of traditional back EMF observers in high-speed sections, achieving optimal motor performance across the entire speed range and improving the system's integrity and adaptability.

[0030] A permanent magnet synchronous motor control system for implementing the above-described method, characterized in that it comprises:

[0031] The microcontroller (MCU) is configured to perform the speed determination, rotor initial positioning, full-order state observer, and control algorithm switching.

[0032] Current sampling circuit, used to collect motor phase current;

[0033] A three-phase inverter bridge drive circuit is used to drive the permanent magnet synchronous motor.

[0034] The MCU obtains current feedback through the current sampling circuit and generates a PWM signal to control the three-phase inverter bridge drive circuit based on the angle and speed estimated by the full-order state observer.

[0035] Preferably, the current sampling circuit is a single-resistor sampling circuit. The MCU samples the current twice within one PWM cycle, collects the current of two phases according to the PWM relationship of each sector, and at any time, the vector sum of the three-phase currents is 0, thus reconstructing the current of the other phase.

[0036] By limiting the use of a single-resistor sampling circuit, it is demonstrated that the high-performance algorithm of this invention can be successfully adapted to the most cost-effective hardware solution. While ensuring excellent silent startup performance, it achieves significant manufacturing cost advantages, breaking the conventional wisdom that high performance inevitably means high cost, and is highly competitive in the market.

[0037] Preferably, the current sampling circuit is a dual-resistor sampling circuit.

[0038] By limiting the use of dual-resistor sampling circuits, this invention demonstrates high flexibility and compatibility, providing better hardware options for application scenarios with extreme requirements for reliability and performance, and broadening the application scope of the technology. Beneficial effects

[0039] Compared with the closest existing technology, the beneficial effects of this invention are significant and multifaceted:

[0040] The root cause of startup noise is reduced: By completely eliminating the core noise-generating stage of "open-loop forced drag" and greatly shortening and optimizing the "pre-positioning" process, this invention eliminates the two largest noise sources in traditional solutions from the control flow perspective. Experiments show that electromagnetic noise and high-frequency howling during startup can be reduced by 10-15 decibels, resulting in a significant improvement in subjective auditory perception.

[0041] Breakthrough in low-speed performance and stability: The proposed full-order state observer method is completely independent of back EMF, thus overcoming the inherent technical bottleneck of traditional methods that cannot accurately observe back EMF signals at zero and extremely low speeds. It achieves truly smooth and stable closed-loop control from zero speed, improving the system's startup reliability under harsh operating conditions.

[0042] Improved dynamic response and control accuracy: Because high-precision closed-loop control is implemented from the initial startup phase, the rotor position estimation error is significantly smaller than the open-loop prediction, resulting in a substantial reduction in torque ripple. This not only reduces noise but also improves startup smoothness and response speed, while minimizing unnecessary energy loss.

[0043] Excellent compatibility and cost-effectiveness: The method described in this invention requires very little hardware modification and can be perfectly adapted to various sampling schemes, from low-cost single-resistor to high-performance dual-resistor. This provides flexible cost-effectiveness options for products with different positioning, greatly enhancing the practicality and market potential of the technology. Attached Figure Description

[0044] Figure 1 : A hardware structure block diagram of a motor control system according to an embodiment of the present invention.

[0045] Figure 2 : Overall flowchart of the control method of the present invention.

[0046] Figure 3 The simulation / experimental waveform diagram showing how the rotor angle observation value quickly converges to the true value after using the method of this invention.

[0047] Figure 4 : A single-resistor sampling three-phase inverter bridge topology circuit diagram applicable to the present invention.

[0048] Figure 5 : A three-phase inverter bridge topology circuit diagram with dual resistance sampling applicable to the present invention. Detailed Implementation

[0049] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0050] Through in-depth analysis and practice, the inventors discovered that existing sensorless starting technologies for permanent magnet synchronous motors (PMSMs) have the following inherent drawbacks:

[0051] The inherent noise sources in the startup process cannot be fundamentally eliminated: In the traditional three-stage startup strategy of "pre-positioning - open-loop forced drag - closed-loop", the large current surge in the pre-positioning stage produces a noticeable electromagnetic "humming" sound; while in the open-loop forced drag stage, because the controller cannot know the true position of the rotor, there is a continuous error between the applied voltage vector and the actual rotor position, resulting in significant torque pulsation and current harmonics, thus causing high-frequency "whistling" and mechanical vibration. The working mechanism of these two stages determines that they are themselves the main noise sources. Under the current technological framework, they can only be improved to a limited extent by optimizing parameters (such as reducing the positioning current and adjusting the V / F curve), and the noise cannot be fundamentally eliminated.

[0052] Insufficient observation capability at extremely low speeds leads to instability during the switching process: Existing technologies rely on the back electromotive force (BEMF) of the motor to estimate rotor position and enter the closed loop. However, in the extremely low-speed region (especially at zero speed and near zero speed), the BEMF signal amplitude is extremely small, resulting in a very low signal-to-noise ratio. This causes the BEMF-based observer to have large estimation errors, slow convergence speed, or even instability at this time. Therefore, the timing of the switch from open loop to closed loop is difficult to control, and the switching process is prone to speed and torque surges, introducing additional noise and even causing start-up failure.

[0053] Balancing control performance and system cost is challenging: Existing technologies face a dilemma in hardware sampling schemes. While dual-resistor sampling schemes offer simple current reconstruction, they are costly; single-resistor sampling schemes are inexpensive but require complex sampling timing and reconstruction algorithms, placing high demands on the computational power of the microcontroller (MCU), and are prone to introducing errors during reconstruction, affecting the accuracy of model-based observers, thus limiting the realization of low-cost, high-performance solutions.

[0054] The inventors discovered that, faced with the aforementioned technical challenges, they did not limit themselves to parameter optimization within the existing three-stage process. Instead, they re-examined and deeply analyzed the root causes of the problem, leading to crucial discoveries:

[0055] Discovery 1: The open-loop forced drag stage is the main source of startup noise. The inventors recognized that the "open-loop forced drag" stage is the core source of noise and vibration throughout the startup process. The fundamental reason for the existence of this stage is to forcibly accelerate the motor before the back electromotive force is strong enough. If a method can be found to accurately estimate the rotor position at extremely low speeds or even zero speeds, it may be possible to bypass or eliminate this core stage that generates noise, and directly switch to closed-loop control after initial positioning.

[0056] Discovery 2: Solving the problem of observation at extremely low speeds requires abandoning the dependence on back electromotive force (EMF). The inventors discovered that to achieve stable starting at extremely low speeds, it is necessary to break the technical bias that "position observation must rely on back EMF." The key lies in finding an observation method that does not depend on back EMF. Through the study of the mathematical model of the motor itself, the inventors realized that the voltage (U), which is represented as a DC signal in a synchronous rotating coordinate system, can be used. d U q ) and current (I) d ,I q ), combined with motor parameters (R) s ,L d ,L q ,Ψ f A full-order state observer based on a motor model is constructed. By solving the motor state equations, this observer can directly estimate the rotor's angular velocity and position in the low-speed domain, thus perfectly avoiding the fatal flaw of weak back EMF signals at low speeds.

[0057] Discovery 3: The initial convergence of the observer is crucial for achieving "direct closed-loop startup." The inventors discovered in experiments that although a full-order state observer theoretically holds promise for solving low-speed observation problems, in the "absolute zero speed" state before the motor actually starts, the observer model may struggle to start or converge slowly due to insufficient excitation. To address this, the inventors creatively proposed a technique of applying a minimum lower limit (e.g., 1%-5% of the rated speed) to the angular velocity estimated by the observer. This measure is equivalent to pre-setting a tiny "virtual speed" within the observer, ensuring that the observer can quickly establish and maintain a stable operating state in the zero-speed stage, thus providing a reliable guarantee for achieving "direct closed-loop startup."

[0058] In summary, the inventors' core discovery lies in the fact that by employing a full-order state observer that does not rely on back EMF and ensuring its rapid convergence near zero velocity, the traditional open-loop drag stage can be skipped, allowing direct entry into closed-loop control after prepositioning, thereby fundamentally optimizing startup noise. This discovery lays the theoretical foundation for the technical solution proposed in this invention.

[0059] Therefore, the present invention provides the following technical solution:

[0060] A method for optimizing the low-speed start-up noise of a permanent magnet synchronous motor blower, characterized by comprising the following steps:

[0061] In response to the start command, determine whether the current speed of the permanent magnet synchronous motor is zero;

[0062] If the current rotational speed is zero, perform an initial rotor positioning operation.

[0063] After the initial rotor positioning operation, without going through the open-loop strong drag stage, the full-order state observer based on the motor model is directly activated for closed-loop control.

[0064] The full-order state observer is based on the motor voltage equation in the synchronous rotating coordinate system. It estimates the rotor's angular velocity and angle in real time according to the input voltage signal and the sampled current signal, and outputs the results to the field-oriented control (FOC) loop.

[0065] Compared to the traditional three-stage start-up method of "pre-positioning-open-loop forced drag-closed-loop", this invention eliminates the open-loop forced drag stage and directly switches to closed-loop control based on model observer after initial positioning. This fundamentally eliminates torque pulsation and high-frequency electromagnetic noise caused by position estimation error in open-loop forced drag, and achieves root-cause noise reduction and stability improvement in the start-up process.

[0066] Specifically, it completely eliminates the significant torque pulsation and high-frequency electromagnetic noise caused by the asynchrony between the voltage vector and the actual rotor position during the "open-loop forced drag" stage. This is a fundamental change to the traditional three-stage startup process, solving the main noise problem at its source.

[0067] Furthermore, since the system operates under closed-loop control from the initial startup stage, it has the ability to provide feedback and correction for the rotor position, making the startup process smoother and avoiding the jitter and shock that may result from open-loop startup.

[0068] This technical solution simplifies the three-stage process into two stages (positioning and closed-loop), reducing the number of control mode switching operations, lowering the risk of failure due to complex switching logic or improper switching timing, and improving the robustness of the system.

[0069] In some embodiments, the rotor initial positioning operation is as follows: injecting a DC current vector of a preset magnitude and direction into the stator winding of the motor, and using the magnetoresistive effect to align the rotor permanent magnet to a determined initial position.

[0070] By limiting the initial positioning to injecting a preset DC current vector, compared with the traditional positioning method that may use a larger current or voltage, a shorter and gentler positioning is achieved, which significantly reduces the current surge and accompanying electromagnetic noise in the initial positioning stage, and provides optimized start-up conditions for subsequent observers.

[0071] Specifically, this technical solution achieves a more "gentle" positioning by controlling the magnitude and duration of the injected current. Compared to traditional pre-positioning methods that may use large voltage / current for a "hard pull," it reduces the current surge and resulting electromagnetic noise during initial positioning. The purpose of this positioning is to provide a rough but accurate initial position estimate for the full-order state observer, rather than precise forced positioning. This positioning strategy works better with the subsequent closed-loop startup of the observer, demonstrating the synergistic optimization of the overall solution.

[0072] In some embodiments, when the full-order state observer is enabled, a lower limit is applied to the estimated angular velocity, which is set to 1% to 5% of the motor's rated angular velocity.

[0073] By setting a lower limit for the angular velocity estimated by the full-order state observer, the theoretical problem of poor observability and difficulty in starting the observer near zero velocity is effectively solved, ensuring the rapid convergence and stability of the observer at the moment of startup, and providing a key guarantee for achieving silent and rapid startup.

[0074] Specifically, this technical solution addresses the theoretical and practical challenges of observers near zero speed. Full-order state observers theoretically face the challenge of deteriorating observability near zero speed. This technique introduces a tiny "virtual speed" to ensure the observer model is in a stable operating state before the motor actually starts, solving the convergence problem when starting from absolute zero speed. This lower limit ensures that even before the rotor rotates, the observer can generate an effective rotating magnetic field estimate, thus quickly and accurately capturing the rotor's position the instant it begins to move slightly. This significantly shortens the transition time from positioning to stable closed-loop operation, providing a crucial guarantee for achieving "silent start-up."

[0075] In some embodiments, the full-order state observer can achieve real-time angle convergence within two electrical cycles after the rotor initial positioning operation is completed, thereby realizing full-velocity domain angle closed-loop control.

[0076] The rapid convergence of the observer within two electrical cycles demonstrates, through quantitative indicators, that the present invention achieves an order-of-magnitude improvement in response speed compared to the traditional open-loop strong drag process that requires tens or even hundreds of electrical cycles. This rapid locking capability directly brings about a qualitative leap in start-up smoothness and noise reduction.

[0077] More specifically, this technical solution quantifies the leap in response speed, a direct and quantitative proof of the superiority of the aforementioned solution. Traditional open-loop forced-drive phases may require tens or even hundreds of electrical cycles to drive the motor to a sufficiently high speed. This invention can complete precise positioning and enter a stable closed loop within an extremely short two electrical cycles, resulting in an order-of-magnitude improvement in response speed. Any control delay during startup translates into torque fluctuations and noise. Such rapid convergence means the motor is almost instantly "locked in" and smoothly accelerates upon receiving the start command, minimizing the duration of unstable states and directly achieving excellent noise reduction.

[0078] In some embodiments, the method further includes:

[0079] If the current speed is not zero, or when the motor operating speed is higher than a preset threshold, switch to closed-loop control based on the back EMF observer.

[0080] By adding a smart switching strategy for high and low speed observers, the control system not only solves the problem of low-speed start-up noise, but also leverages the advantages of traditional back EMF observers in high-speed sections, achieving optimal motor performance across the entire speed range and improving the system's integrity and adaptability.

[0081] More specifically, this invention creatively solves the low-speed problem without excluding mature high-speed solutions. By employing the most suitable observer in different speed ranges (a model observer for low speed and a back EMF observer for high speed), it ensures that the motor operates in an optimal state throughout the entire speed range from zero to high speed, combining the advantages of quiet operation at low speeds and high efficiency at high speeds. This allows the system to adapt to a wider range of working scenarios, handling both extremely quiet low-speed starts and high-efficiency high-speed operation with ease, thus enhancing the overall competitiveness of the product.

[0082] The method specifically includes the following steps:

[0083] S1: Status judgment. When the motor is powered on and started, the MCU control unit receives the speed command and determines whether the current motor speed is zero.

[0084] S2: Initial positioning (if the speed is zero). If S1 determines yes, then perform a rotor initial positioning operation: inject a preset DC current vector into the stator winding (e.g., set the d-axis current I). d =0, q-axis current I q =I start I start (For a small, constant value), a stationary directional magnetic field is generated. Under the action of reluctance torque, the rotor permanent magnet will rapidly align to the direction of this magnetic field (e.g., 90° electrical angle). This process lasts for a very short time (typically 5-20 ms) and is intended to provide the observer with a coarse initial position estimate.

[0085] S3: Directly switch to closed-loop low-speed control. After positioning, do not enter open-loop strong drag mode, but immediately activate the full-order state observer. This observer operates in a synchronous rotating coordinate system (dq axis), and its input is the given voltage command U. d U q and the current I from the sampling feedback d ,I q The observer constructs a state-space model based on the voltage equations of the permanent magnet synchronous motor (as shown below), and directly estimates the rotor's angular velocity ω through real-time calculations. r .

[0086]

[0087] (where R) s L is the stator resistance. d ,L q For direct-axis and quadrature-axis inductors, Ψ f (Permanent magnet magnetic flux).

[0088] S4: Angle acquisition and velocity assurance. For the estimated angular velocity ω... rIntegrating, we obtain the real-time rotor angle θ. To ensure the stability and fast convergence of the observer near zero velocity, we adjust ω... r Apply a positive lower limit value (e.g., limit its minimum to 1%-5% of the motor's rated angular velocity). This measure ensures that even when stationary and at extremely low speeds, there is an effective "rotational" excitation within the observer model, enabling it to converge quickly and accurately identify the true rotor position in a very short time (typically less than 2 electrical cycles).

[0089] S5: Closed-loop feedback and mode switching. The estimated ω... r The current and speed loops are fed back to the FOC loop in real time to complete closed-loop control. If the motor's current speed is not zero, or the operating speed is higher than a preset threshold (e.g., 800 rpm), the control switches to a high-speed observer based on back EMF to achieve optimal performance across the entire speed range.

[0090] Supporting System: This invention also provides a permanent magnet synchronous motor control system for implementing the above-described method, comprising:

[0091] The microcontroller (MCU) is configured to perform the speed determination, rotor initial positioning, full-order state observer, and control algorithm switching.

[0092] Current sampling circuit, used to collect motor phase current;

[0093] A three-phase inverter bridge drive circuit is used to drive the permanent magnet synchronous motor.

[0094] The MCU obtains current feedback through the current sampling circuit and, based on the angle estimated by the full-order state observer, assesses the technical effect relative to existing technologies.

[0095] In some embodiments, the current sampling circuit is a single-resistor sampling circuit. The MCU samples the current twice within one PWM cycle, collects the current of two phases according to the PWM relationship of each sector, and at any time, the vector sum of the three-phase currents is 0, thus reconstructing the current of the other phase.

[0096] By limiting the use of a single-resistor sampling circuit, it is demonstrated that the high-performance algorithm of this invention can be successfully adapted to the most cost-effective hardware solution. While ensuring excellent silent startup performance, it achieves significant manufacturing cost advantages, breaking the conventional wisdom that high performance inevitably means high cost, and is highly competitive in the market.

[0097] More specifically, while single-resistor sampling is the lowest-cost hardware solution, its reconstruction algorithm is complex and poses challenges to control performance. This invention successfully combines an innovative software algorithm (full-order state observer) with a lowest-cost hardware solution, breaking the traditional notion that "high performance inevitably means high cost." It provides a highly competitive solution, demonstrating that the control algorithm of this invention can overcome the inherent difficulties of single-resistor sampling and operate stably. This, in turn, illustrates the robustness and advanced nature of the algorithm itself.

[0098] In some embodiments, the current sampling circuit is a dual-resistor sampling circuit.

[0099] By limiting the use of dual-resistor sampling circuits, this invention demonstrates its high flexibility and compatibility, providing a superior hardware option for applications with stringent reliability and performance requirements, and broadening the scope of its applications. For applications that prioritize ultimate reliability and performance regardless of cost (such as high-end vehicles), a simpler dual-resistor scheme can be employed to further reduce software complexity and potential computational errors, ensuring optimal system performance.

[0100] The supporting system specifically includes:

[0101] MCU control unit: As the control core, it has a built-in program memory to store and execute the speed judgment, initial positioning, full-order state observer and FOC algorithm.

[0102] Current sampling module: Used for high-precision acquisition of motor phase current. It can preferably be a single-resistor sampling topology (by matching the PWM output logic of each sector, two phase currents are obtained through two samplings, and the third phase current is calculated from the two phase currents to reduce costs) or a dual-resistor sampling topology (by measuring the two phase currents to calculate the third phase, simplifying the algorithm).

[0103] Three-phase inverter bridge drive circuit: typically a B6 bridge structure, consisting of six power switching transistors (such as MOSFETs or IGBTs), receiving PWM signals generated by the MCU to drive the permanent magnet synchronous motor.

[0104] Signal conditioning and protection circuitry includes power management, signal isolation, and overcurrent / overvoltage protection to ensure reliable system operation. Example 1: Low-cost vehicle-mounted blower control based on single-resistor sampling

[0105] Reference Figure 1 and Figure 4The hardware of this control system includes: vehicle-mounted 12V / 24V power supply (KL30), fuse and power protection circuit, MCU based on ARM Cortex-M core (such as ST Microelectronics' STM32F3 series), single resistor current sampling circuit (including operational amplifier and ADC), three-phase full-bridge inverter (composed of 6 N-MOSFETs) and permanent magnet synchronous blower motor.

[0106] Specifically, Figure 4 The system consists of three half-bridges composed of six switching transistors (MOSFETs / IGBTs). Each half-bridge drives one phase of the motor. The rotation of the motor is controlled by the regular conduction of the upper and lower switching transistors. Finally, a three-phase sinusoidal current waveform is measured on the PMSM blower.

[0107] Software process (refer to) Figure 2 ):

[0108] 1. After the system powers on and initializes, the MCU determines the current motor speed. If it is 0 rpm, the startup process begins.

[0109] 2. Initial Positioning: The MCU outputs PWM, injecting a constant current I of 10% of the rated current into the q-axis. q This process lasts for approximately 10 ms, stabilizing the rotor at an electrical angle of around 90°.

[0110] 3. Direct Closed-Loop Startup: After localization is complete, the full-order state observer module is immediately invoked. This observer uses the U calculated by the FOC loop. d U q and the sampled and reconstructed I d ,I q For input. The MCU calculates the motor parameters (R... s ,L d ,L q ,Ψ f Solve the observer equations in real time to estimate ω. r For ω r Perform a lower limit (e.g., limit the minimum value to 50 rpm), and then integrate to obtain θ.

[0111] 4. FOC closed-loop operation: estimated θ and ω r The feedback is immediately sent to the Park / inverse Park transform module and the speed PI regulator to complete the full FOC control. The observer can converge quickly within 2 electrical cycles (approximately a few milliseconds), such as... Figure 3 As shown, this enables silent startup.

[0112] 5. If the target speed is higher than 800 rpm, when the actual speed reaches this threshold, smoothly switch to the sliding mode observer (SMO) based on back electromotive force for high-speed control. Example 2: High-performance application based on dual-resistor sampling

[0113] The difference from Example 1 is that the current sampling module adopts... Figure 5 The diagram shows a dual-resistor topology (with sampling resistors placed in the lower arms of phases A and B respectively). This scheme estimates the third current parameter by collecting the current parameters of two phases, thus reconstructing the motor phase current parameters. The logic is simpler than that of a single resistor, but the cost is higher. It eliminates the need for complex single-resistor sampling timing control, and the current reconstruction algorithm is simpler and more direct, helping to reduce CPU load and improve system robustness, making it suitable for applications with higher reliability requirements.

[0114] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for optimizing the low-speed start-up noise of a permanent magnet synchronous motor blower, characterized in that, Includes the following steps: In response to the start command, determine whether the current speed of the permanent magnet synchronous motor is zero; If the current rotational speed is zero, perform an initial rotor positioning operation. After the initial rotor positioning operation, without going through the open-loop strong drag stage, the full-order state observer based on the motor model is directly activated for closed-loop control. The full-order state observer is based on the motor voltage equation in the synchronous rotating coordinate system. It estimates the rotor's angular velocity and angle in real time according to the input voltage signal and the sampled current signal, and outputs the results to the field-oriented control (FOC) loop.

2. The method according to claim 1, characterized in that, The rotor initial positioning operation is as follows: inject a DC current vector of preset magnitude and direction into the stator winding of the motor, and use the magnetoresistive effect to align the rotor permanent magnet to a determined initial position.

3. The method according to claim 2, characterized in that, The full-order state observer can achieve real-time angle convergence within 2 electrical cycles after the rotor initial positioning operation is completed, thereby realizing full-speed domain angle closed-loop control.

4. The method according to claim 3, characterized in that, The full-order state observer is able to converge and accurately identify the real-time angle of the rotor within two electrical cycles after the rotor initial positioning operation is completed.

5. The method according to claim 1, characterized in that, The method further includes: If the current speed is not zero, or when the motor operating speed is higher than a preset threshold, switch to closed-loop control based on the back EMF observer.

6. A permanent magnet synchronous motor control system for implementing the method of any one of claims 1-5, characterized in that, include: The microcontroller (MCU) is configured to perform the speed determination, rotor initial positioning, full-order state observer, and control algorithm switching. Current sampling circuit, used to collect motor phase current; A three-phase inverter bridge drive circuit is used to drive the permanent magnet synchronous motor. The MCU obtains current feedback through the current sampling circuit and generates a PWM signal to control the three-phase inverter bridge drive circuit based on the angle and speed estimated by the full-order state observer.

7. The control system according to claim 6, characterized in that, The current sampling circuit is a single-resistor sampling circuit. The MCU performs two current samplings within one PWM cycle and reconstructs the three-phase current using the principle that the vector sum of the three-phase currents is 0 at any given time.

8. The control system according to claim 6, characterized in that, The current sampling circuit is a dual-resistor sampling circuit.

Images